Despite efficient suppression of plasma viremia in HIV-infected patients on cART, replication competent virus is still recoverable from a variety of anatomic sites. A major reservoir of HIV and pathogenic SIV infection in rhesus macaques (RM) are lymph node follicles (via replication in TFH cells and viral trapping by FDC) that generally prohibit access to cytotoxic CD8+ T cells. Interestingly, nonpathogenic SIV infection of African green monkeys (AGM) exhibits high viral loads in blood and the gut mucosae, but lacks virus in lymphoid follicles. Natural killer (NK) cells provide rapid early responses to HIV/SIV infections and contribute substantially to disease modulation and vaccine protection, but the underlying mechanisms for NK cell surveillance and modulation of HIV/SIV reservoirs is unclear. Our recent research indicates that SIV can be effectively cleared from lymphoid follicles in nonpathogenic AGM but not pathogenic RM hosts and is dependent on an IL-15- dependent redistribution of NK cells to lymph node follicles and increased CXCR5 expression on AGM NK (Huot et al. Nature Medicine 2017). In this innovative proposal we will utilize three SIV-infected nonhuman primate models to test the central hypothesis that NK cells critically contribute to the reduction of SIV reservoirs. Specifically we will: (1) Determine mechanisms associated with NK cell-mediated control of SIV reservoirs; and (2) Determine mechanisms associated with NK cell failure to control SIV reservoirs in lymph nodes and associated pathogenesis. Application of these data could contribute substantially to future HIV/SIV vaccines, immunotherapeutics, or reservoir purging strategies that harness the potent antiviral potential of NK cells.